Upscaling permeability anisotropy in digital sandstones using convolutional neural networks

Najafi, Atefeh; Siavashi, Javad; Ebadi, Mohammad; Sharifi, Mohammad; Fahimpour, Jalal; Koroteev, Dmitry

doi:10.1016/j.jngse.2021.104263

Cited by 10 publications

(2 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, using 3D data volume as input, although CNN has good accuracy and relatively low computational cost compared to numerical methods, CNN still requires high computational costs during training (Siavashi et al., 2022). Meanwhile, there is a contradiction that the resolution needs to be sacrificed to expand the scanning field of view (Najafi et al., 2021; Siavashi et al., 2022). In this regard, 2D slices are cheaper and more available experimentally, compare with 3D structured data volumes.…”

Section: Introductionmentioning

confidence: 99%

Predicting 3D Physical Properties From a Single 2D Slice Based on Convolutional Neural Networks: 2D‐Slice‐To‐3D‐Properties for Porous Rocks

Chen,

Yang,

et al. 2023

Water Resources Research

View full text Add to dashboard Cite

The 3D physical properties of porous rocks directly determine the subsurface flow and modelling. However, predicting a wide range of 3D physical rocks remains a formidable challenge and requires a large amount of input. Reliable microstructure‐property correlations can accurately predict 3D physical properties, avoiding time‐consuming experimental testing. Here, we propose a new dimensionless CNN‐based method to find inherent correlations for accurately predicting 3D properties of porous rocks using only a single 2D slice or a series of 2D slices. Training and testing were conducted on 9 properties of 16,409 semi‐realistic 3D porous rocks. Eight properties, except for the formation factor, were predicted with an acceptable correlation coefficient of 0.92. Furthermore, the modeling results indicate that the network architecture of the model has a significant impact on its performance. The classical CNN Model 4 achieved an average validating loss of 1.0×10‐3. Interval 2D slice sampling can significantly reduce the computational cost in training model. The developed model was verified and tested by five independent porous samples. Additionally, the proposed method accurately predicted permeability in different directions using only 2D slices trained in one direction, for porous rocks where anisotropy is not significant. This work demonstrates the ability to predict various 3D properties rapidly and accurately using only a single 2D slice from semi‐realistic 3D volumes. It provides insights into the conversion of 2D slices to 3D properties for porous media in a wide range of applications, including property estimation from 2D images without structural construction and simulation in 3D.This article is protected by copyright. All rights reserved.

show abstract

Section: Introductionmentioning

confidence: 99%

Predicting 3D Physical Properties From a Single 2D Slice Based on Convolutional Neural Networks: 2D‐Slice‐To‐3D‐Properties for Porous Rocks

Chen,

Yang,

et al. 2023

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…The generated results of PSS can be questionable due to the trade‐off between the spatial resolution and FOV of μ xCT images (Najafi et al., 2021; Yue et al., 2016). The addressed trade‐off can result in failure to capture pore sizes less than the spatial resolution and consequently gives rise to uncertainty of pore geometry (Blykers et al., 2021; Saxena, Hofmann, Alpak, Berg, et al., 2017).…”

Section: Introductionmentioning

confidence: 99%

Predictive Soft Computing Methods for Building Digital Rock Models Verified by Positron Emission Tomography Experiments

Ebadi

Armstrong

Mostaghimi

et al. 2022

Water Resources Research

Self Cite

View full text Add to dashboard Cite

The influence of core‐scale heterogeneity on continuum‐scale flow and laboratory measurements are not well understood. To address this issue, we propose a fully automated workflow based on soft computing to characterize the heterogeneous flow properties of cores for predictive continuum‐scale models. While the proposed AI‐based workflow inherently has no trained knowledge of rock petrophysical properties, our results demonstrate that image features and morphological properties provide sufficient measures for petrophysical classification. Micro X‐ray computed tomography (μxCT) image features were extracted from full core plug images by using a Convolutional Neural Network and Minkowski functional measurements. The features were then classified into specific classes using Principal Component Analysis followed by K‐means clustering. Next, the petrophysical properties of each class were evaluated using pore‐scale simulations to substantiate that unique classes were identified. The μxCT image was then up‐scaled to a continuum‐scale grid based on the defined classes. Last, simulation results were evaluated against real‐time flooding data monitored by Positron Emission Tomography. Both homogeneous sandstone and heterogeneous carbonate were tested. Simulation and experimental saturation profiles compared well, demonstrating that the workflow provided high‐fidelity characterization. Overall, we provided a novel workflow to build digital rock models in a fully automated way to better understand the impacts of heterogeneity on flow.

show abstract

Relative permeability estimation using mercury injection capillary pressure measurements based on deep learning approaches

Duan,

Kang,

Deng

et al. 2024

J Petrol Explor Prod Technol

View full text Add to dashboard Cite

Relative permeability (RP) curves which provide fundamental insights into porous media flow behavior serve as critical parameters in reservoir engineering and numerical simulation studies. However, obtaining accurate RP curves remains a challenge due to expensive experimental costs, core contamination, measurement errors, and other factors. To address this issue, an innovative approach using deep learning strategy is proposed for the prediction of rock sample RP curves directly from mercury injection capillary pressure (MICP) measurements which include the mercury injection curve, mercury withdrawal curve, and pore size distribution. To capture the distinct characteristics of different rock samples' MICP curves effectively, the Gramian Angular Field (GAF) based graph transformation method is introduced for mapping the curves into richly informative image forms. Subsequently, these 2D images are combined into three-channel red, green, blue (RGB) images and fed into a Convolutional Long Short-Term Memory (ConvLSTM) model within our established self-supervised learning framework. Simultaneously the dependencies and evolutionary sequences among image samples are captured through the limited MICP-RP samples and self-supervised learning framework. After that, a highly generalized RP curve calculation proxy framework based on deep learning called RPCDL is constructed by the autonomously generated nearly infinite training samples. The remarkable performance of the proposed method is verified with the experimental data from rock samples in the X oilfield. When applied to 37 small-sample data spaces for the prediction of 10 test samples, the average relative error is 3.6%, which demonstrates the effectiveness of our approach in mapping MICP experimental results to corresponding RP curves. Moreover, the comparison study against traditional CNN and LSTM illustrated the great performance of the RPCDL method in the prediction of both So and Sw lines in oil–water RP curves. To this end, this method offers an intelligent and robust means for efficiently estimating RP curves in various reservoir engineering scenarios without costly experiments.

show abstract

Upscaling permeability anisotropy in digital sandstones using convolutional neural networks

Cited by 10 publications

References 63 publications

Predicting 3D Physical Properties From a Single 2D Slice Based on Convolutional Neural Networks: 2D‐Slice‐To‐3D‐Properties for Porous Rocks

Predicting 3D Physical Properties From a Single 2D Slice Based on Convolutional Neural Networks: 2D‐Slice‐To‐3D‐Properties for Porous Rocks

Predictive Soft Computing Methods for Building Digital Rock Models Verified by Positron Emission Tomography Experiments

Relative permeability estimation using mercury injection capillary pressure measurements based on deep learning approaches

Contact Info

Product

Resources

About